The Art of Oilseed Meal Grinding * G E O R G E R. T H O M A S , Vice President , Pra te r Industr ies , Inc., Chicago, Il l inois 60650

Abstract Meal grinding must be set up to give a definite and

desired product. The original concept of solvent ex- tracted soybean meal grinding was to use the two- stage grinding system. However, the advent of front end hull removal equipment resulted in the develop- ment of the single stage grinding system, utilizing the grinders with just one sifter. This system lends itself well to grinding on stream. Meal grinding installations must be carefully planned to accommodate all parameters, such as desired consistency of finished product, dust control, conveyance systems, placement of all equipment in meal grinding system, and pro- visions for hot-meal, or wet-meal grinding require- ments.

The operation of meal grinding, whether it be for soybean or other oil-bearing seeds, is best looked at in two main areas, the mechanics of the grinding device, and the placing of the mill in the processing system.

Changes in the preparat ion of meal pr ior to the grinding process have caused changes in the effectiveness of meal grinding. Perhaps the biggest and most significant change was the change from the expeller cake process to the solvent extraction process. Also, in later years, the advent of front end dehulling equipment greatly facilitated the processing of meal through the grinding process.

The problem, therefore, becomes one of selecting the proper type of pulverizer to meet the objectives and specifications of a finished product of the oilseed meal industry. Although in terms of final granulation the specifications may vary somewhat, they really are very close. One major feed company has the following speci- fications for their finished meal grinding: Everything through a 10 mesh screen, with 50% maximum through a 24 mesh screen. Another company gave an older original specification of approximately 11% moisture meal of 0% on 10 mesh, 10%-20% on 20 mesh, 40%-60% on 60 mesh, 25%-40% on 100 mesh, 2% maximum through 100 mesh. The companies have since revised the figures because these seemed to be just too fine. In general i t can be safely assumed that the finished meal specification for most ap- plications calls for no more than 1% on a 10 mesh screen and no more than 1% through an 80 mesh screen.

Because of this specification, the selection of a mill which will not grind the friable meal too fine is important. The first important parameter to achieve this condition is to arrive at the proper relationship between horsepower and screen area. The nature of grinding a product like solvent extracted soybean meal is such that a mill with a large screen area is highly desirable. One cannot simply add more horsepower to an existing small screen mill to achieve a good high volume meal grinder, because as the driving force (horsepower applied) is increased, while holding to a fixed screen area, the resultant grind will be finer. Obviously in meal grinding this is not desirable. The part icular mill which is grinding most of the soybean meal processed in the United States today has not one, but two screens in it, giving a total of 1,710 in? of screen area, with a driving force of 100 horsepower or a ratio of over 17 i n / o f screen area per horsepower applied.

A second important consideration is to get the product being ground out of the mill, after it has been reduced to the proper granulation, as fast as possible. Of course, a large screen area will help to do this. In addition to a the mill is also of help. A pulverizer, which is really screen area, however, proper use of the air passing through nothing more than a rotor with swinging hammers turning inside a mill case or an enclosure, can be considered as a

1One of nine papers to be published from the Symposium on Solvent Extraction Techniques for Soybean and Other Seeds, presented at the AOCS Meeting, Minneapolis, October, 1969.

fan. The ai r created by the turning of the rotor at 1800 rpm, par t icular ly in a large diameter mill, is definitely a reality and, proper ly used, can be an asset to the grinding of meal.

The air natural ly enters a t the eye of the mill through both sides, under the suction caused by the turning of the rotor, and exits at the bottom of the discharge. I f this system were to be choked or starved for air in any way, the capacity and resultant granulation would be adversely affected. In other words, it is a good policy to let the mill "breathe," and i f the discharge goes into a mechanical conveying system, it is sufficient to balance the air system with a small dust collecting system sized to accommodate the air generated by the mill itself. This is normally done by taking suction from a plenum chamber built on top of the conveyor further on down the line, with the return of separated particles from the collector to the system. In setting up the air relief system in this manner, you should have about 1~ in. water pressure (negative) at the discharge of the mill. Referring again to the 100 t I P mill mentioned previously, the air relief on this mill amounts to approximately 2100 cfm.

In an effort to closely control granulation, a third consideration is important. Gradual reduction, rather than instant reduction, will greatly reduce fines. Our meal grinder uses the Triple Reduction process, rather than the instantaneous confrontation o£ the product into the hammer section of the mill. The advent of side feeding also has the advantage of spreading out the product being ground, so as to put the meal across the total screen area for final reduction in the most effective manner possible. Consider for a moment a woman preparing mashed potatoes. She first cuts up the potato in smaller pieces before cooking and final mashing. She does not at tempt to mash the whole potato. This gradual reduction, or Triple Reduction has also gained much prominence in the distilling industry, where controlled granulation without excessive fines on corn, rye and malt products is very important.

The meal-grinding stage in a meal production plant comes usually af ter the extractor, the desolventizer-toaster, the dryer ( if possible), and the meal cooler ( if possible). I f the dryer is not installed or is placed after the grinding system, the condition is known as "wet meal grinding." I f the meal is ground on stream without the benefit of a cooler, the process is known as hot meal grinding. The meal usually has about 18-20% moisture from the de- solventizer-toaster and i t is desirable to get the moisture level down before grinding.

In the early days of solvent extraction soybean opera- tions the most common soybean milling setup was what was known as the two-stage grinding system. In this sys- tem the processed meal from the D.T. tank is first sifted through approximately a No. 12 screen. Some of the meal (sometimes as much as 30%) from the D.T. tank is minus 12 mesh and will pass through the screen into the system without going to the mill. The overs from the first sifter are then ground in as many mills as necessary to keep up with plant production requirements.

I f the product is friable and without hulls i t is pos- sible to operate the first mills at 1200 rpm to reduce the impacting force, and by so doing, reduce the potential of making fines.

The material then passes from the first stage mills, to a second sifter, and into the finished meal conveying system. The overs from the second sifter, which in the early days before front end dehulling consisted of mostly hulls, are then passed through a second stage mill, this time a 1800 rpm mill to insure that all of the material will be reduced to proper size.

The most common system of meal grinding today is (Continued on page 380A)

360A a. AM. On, C~EMISTS' Soc., AI~C,'VST 1970 (VOrJ. 47)

• Oilseed Meal (Continued from page 360A)

known as the single s tage g r ind ing system. The adven t of f r o n t end dehul l ing equ ipmen t makes this system even more popu la r . I n the single stage system, the incoming meal is first passed over a s if ter , and the ma te r i a l t h a t is a l ready down to size passes into the conveying s t ream. The overs f r o m the s i f t e r then are passed to as m a n y mills as necessary to keep u p wi th p l a n t p roduc t ion require- ments . A f t e r g r ind ing , the finished mate r ia l is r e tu rned to t h e or ig ina l s i f t e r and the overs are passed t h r o u g h the mills once again. Usual ly the amoun t of overs r e tu rned to the mills f o r g r i n d i n g a second time, r a r e ly exceeds 20%. These mills a re usua l ly 1800 r p m pulver izers in th is system.

Most of the pulver izers suppl ied to the i ndus t ry today are e i ther rep lacements or addi t ions to exis t ing systems, r a t h e r t h a n newly cons t ruc ted p l an t s or facili t ies. As such, i t is more l ikely to find modifications and addi t ions to e i ther the s ingle or two-s tage g r ind ing systems, r a t h e r t h a n the s imple flow p a t t e r n s suggested here. However , when new faci l i t ies are p l a n n e d today, they p r o b a b l y will include the extractor , desolvent izer- toaster , d rye r and cooler, and then the single s tage on-s t ream g r ind ing set up, as outl ined.

Gr ind ing systems in soybean p l a n t s today are usual ly ope ra t ed on a 24 h r / d a y basis and the capac i ty of the pu lve r i z ing equ ipment usua l ly allows fo r a very com- fo r t ab le g r i n d i n g rate . G r i n d i n g facil i t ies should not opera te a t 100% ful l load capaci ty , so t h a t ammete r read- ings would no t have to be cons tan t ly watched to p reven t overloading. I t is f a r be t t e r to assume a g iven capaci ty based on 7 0 % to 80% ful l load readings . B y so doing the equ ipmen t can be pe rmi t t ed to r u n cont inuous ly with- out problem. Typica l of some of these safe, very con- serva t ive g r i n d i n g ra tes g iv ing an acceptable finished p roduc t are as fol lows: Solvent ex t rac ted soybean meal, 3 0 0 - 4 0 0 / h p / h r ; solvent ex t rac ted cot tonseed meal, 275 l b / h p / h r ; solvent ex t rac ted l inseed meal, 300 l b / h p / h r .

The process ing of soybeans by pulver izers , pa r t i cu la r ly t h r o u g h the scientific pa r t i c l e reduc t ion phase, has given our c o m p a n y a n o p p o r t u n i t y to con t r ibu te to this p h e n o m e n a l success story. W e are h a p p y to have been able to con t r ibu te our p a r t to this most dynamic oilseed process ing indus t ry .

[Received January 2, 1970]

air.~ondltioned roome wttl~ TV. Superb food.

Blolmmi#g Pool, Sol~dum. ~ingIes from $ 14 Doubles from $ ! 8 Lexinglon Ave., 48-49th Sis.

b New York City 10017 TEL: (212) PLaza 5.4000

-- or your travel agent--

FREE S W I M M I N G i n

N2 .C.

ABSTRACTS: BIOCHEMISTRY AND N U T R I T I O N

(Continued from page 379A)

QUANTITATIVE RELATIONSHIP BETWEEN AMOUNT OF DIETARY FAT AND SEVERITY OF ALCOHOLIC FATTY LIVER. C. S. Lieber and L. M. DeCarli (Section of Liver Dis. and Nutr., Bronx Veterans Admin. Hosp., New York, N.Y.). Am. J. Cli/n. Nutr. 23, 474-8 (1970). To assess the quanti tat ive relationship between fa t content of the diet and lipid accumulation in the liver af ter alcohol ingestion, rats were given various isocaloric liquid diets, containing 18% of total calories as protein and 36% as ethanol or isocaloric carbohydrate. The remainder of the calories consisted of varying amounts of f a t (2, 5, 10, 15, 25, 35 or 43% of total calories) and corresponding amounts of carbohydrate. Dietary f a t consisted of ethyl linoleate (2% of total calories), to avoid essential fa t ty acid deficiency, and an olive-corn oil mixture. Af ter 24 days of ethanol and 43% of calories at fa t , hepatic triglycerides increased seven- to elghtfold. With 35% of calories as fat , the increase was five- fold and with 25%, only two- to threefold. No significant decrease in hepatic lipid accumulation was achieved by fur ther reduction in the dietary f a t ; a diet with 25% of calories as f a t (about hMf that of the average United States diet) appears, therefore, to be optimal for minimizing the steatogenie effects of ethanol. An excess of dietary protein did not affect the ethanol-induced steatosis, and even a combination of a high proteindow f a t diet did not achieve full protection against the ethanol-induced hepatic deposition of lipids in the liver. The ethanol effect persisted unchanged for periods up to 3-5 months.

HEPATIC LIPID METABOLISM I1~ HYPOPHYSECTOMIZED AND GROWTH HORMONE-TREATED HYPOPHYSECTOMIZED RATS. J. P. Liberti, R. S. Navon, E. S. Longman and P. F. Jezyk (Dept. of Biochem., Medical College of Virginia, Health Sci. Div., Virginia Commonwealth Univ., Richmond, Va. 23219). Proc. Soe. Exp. Biol. Med. 133, 1346-50 (1970). Changes which occur in hepatic l ipid metabolism of hypophysectomlzed and bovine growth hormone-treated hypophysectomized rats were studied. Synthesis of lipids, as measured by the abili ty of tissue slices to incorporate 8H-glycerol into lipids, decreased rapidly following hypophysectomy. Three weeks af ter hypo- physectomy, glycerol incorporation was approximately 40% of tha t in normal rats and remained depressed over the ex- perimental period. Synthesis of acidic lipids was not ap- preciably affected by hypophysectomy but labeling of all other classes of lipids, part icularly triglycerides, was diminished.

LIPASE IN PANCREAS AND INTESTINAL CONTENTS OF CHICKENS FED HEATED AND RAW SOYBEAN DIETS. S. Lepkovsky and F. Furu ta (Dept. of Poultry Husbandry, Univ. of Calif., Berkeley, Calif. 94720). Poultry Sci. 49, 192-98 (1970). Lipase activity was measured in : pancreases of intact chickens and in pancreases of chickens with ileostomles; intestinal contents from intestines of intact and operated chickens; intestinal contents voided through an ileostomy; and cecal contents from intact chickens. Af te r one feeding of raw soybean diet, the level of lipolytie activity in the pancreases of the chlekeus was reduced to about one-half of tha t found at fast ing, and was in marked contrast to the small decreases of llpolytic activity in the pancreases of chickens fed heated soybean diet.

AORTIC RUPTURE, BODY WEIGHT~ AND BLOOD PRESSURE IN THE TURKEY AS INFLUENCED BY STRAIN, DIETARY FAT, BETA-A!~INO- PROPIONOTRILE FU~[ARATE AND DIETHYLSTILBESTROL. lJ. M. Krista, P. E. Waibel, J . H. Sautter and R. N. Shoffner (Depts. Animal Sci. and Vet. Pathol., Univ. of Minn., St. Paul, Minn. 55101). Poultry Sci. 48, 1954-60 (1969). Three strains of Broad White male turkeys were utilized to deter- mine the effects of dietary fat , beta-aminopropionitrile ( B A P N ) , and diethylstilbestrol (DES) on blood pressure (BP) , incidence of aortic rupture (AR), and body weight (BW). While 10% of animal f a t increased BW significant]y i t did not alter blood pressure or provoke aortic rupture in an experiment of low natural incidence. BAPN at 0.01% did not influence these measures but a t 0.015% decreased BW slightly, increased incidence of AR, and did not affect BP. DES did not al ter BW but i t resulted in reduced B P and greater incidence of AR. Three strains of turkeys were used. Two strains showing greater body weight gains and higher blood pressure also were afflicted to a greater extent by aortic rupture as compared to the third strain. One of the higher incidence strains showed a greater susceptibility to aortic rupture as induced by DES. With one generation of selection for higher and lower blood pressures, the incidence of na tura l aortic rupture was approximately twice as high in the high lines of all three strains.

3 8 0 A J. A~L OIL CI~EMIS~S' Coo., AtrausT 1970 (VET.. 47)